Event-Based Camera Velocimetry Field Reconstruction
University of Pittsburgh researchers have developed a novel method for full-field, non-intrusive measurement of fluid velocity fields using Event-Based Cameras (EBC) combined with a proprietary algorithm. This innovative approach leverages the high spatial and temporal resolution of event-based imaging to generate sparse and scattered velocity information. The developed algorithm processes this data using reduced order modeling and machine learning, reconstructing a dense flow field with high accuracy. This technology offers a cost-effective and efficient solution for real-time velocimetry measurements, overcoming limitations of traditional techniques.
Description
The invention utilizes Event-Based Cameras to capture fluid velocity fields at high spatial and temporal resolution. Unlike conventional velocimetry techniques, EBC does not require expensive high-powered lasers. The data obtained from EBC is processed using a novel algorithm that learns the inherent spatio-temporal correlation in the data, reconstructing a dense flow field. This method provides a continuous description of the measured velocity field, enabling faster and more reliable post-processing. The event cameras are significantly cheaper than frame-based cameras, making this technology a robust and affordable solution for fluid flow measurement.Applications
• Fluid velocity measurement• Real-time velocimetry
• Laboratory research tool
• Industrial fluid dynamics